Version 0.49.6
MacOS 15.4.1
Django app, models.py file with about 8k lines and many models.
Selected a model method, about 50 lines of code in the middle of the file.
Added to chat and asked it if it could be optimized.
AI suggested changed, I hit accept.
It began from the top of the file to find the code.
Updated the code, but in doing so I ended up with duplicate models. Have not tried to repeat.
I like the AI, it has good suggestions but I would prefer to make the changes manually.
What settings can I change so that instead of trying to update the code, it just looks at what I give it and then suggests changes, rather than trying to make them.
This only seems to be a problem with large files, and seems due to how it starts at the beginning of a file when the code is 4k lines from the top.
Also, the AI does’t seem to know Cursor itself very well.
Will not stop me from using it, the AI is very helpful, but if you could have some sort of setting related to this, maybe if the file is larger than x lines do not offer to update it?
Also noticed that if I cut/paste from one file to another, then save, it will make some changes after save. I don’t have any auto formatting turned on, and the AI couldn’t figure out what was happenind, but I disabled the Cursor Tab and it stopped happening.
This happened doing a copy/paste between two cursor windows of different projects.
No freezing, performance is not super fast but fine. No show stoppers but the first issue was disturbing as it didn’t hightlight all the duplicate code that occurred.
